Funded Interregional Project Networks
Projects rated A “excellent” and funded:
“Open-Ended Learning for Interactive Robots”
- Project coordinator: Justus Piater, University of Innsbruck
- Project partner: Free University of Bozen-Bolzano, Angelika Peer
- Project partner: University of Trento, Niculae Sebe
Project Title: OLIVER – Open-Ended Learning for Interactive Robots
Content: Who would not like to have a helper robot that does household chores while the family is at work or pursuing recreational activities? For the vision of multi-purpose helper robots to come true, a key open problem is robot skill and knowledge acquisition. We would like to be able to teach robots to perform a great variety of tasks, including tasks in collaboration with humans, and tasks not specifically foreseen by its designers. This poses challenges to machine learning methods, as it must be possible to teach the robot using minimal instruction. Thus, a central problem the robot needs to solve is to understand which aspects of a demonstrated task are crucially important and which can be neglected or are interchangeable. Such understanding allows a robot to perform a task robustly even if the scenario and context are changing, to adapt and alter its strategy if necessary, to judge its success, to recognize its own limitations, and to ask for help in specific ways. Moreover, understanding the essence of a demonstrated task allows the robot to infer the human intent and task progress with respect to an identified goal, enabling it to share the task with humans, offer help or ask for help, resulting in natural human-robot cooperative behavior.
Central Hypotheses: Simple gestural and verbal cues can allow a general-purpose robot to acquire diverse skills while requiring much less human teaching effort than without these cues. The extra information conveyed via these cues allows the robot to engage in natural interaction with humans, e.g. by asking for help or offering help
Methods: A human teaches the robot by moving its arm, adding explicit cues by pointing and other gestures and by simple verbal commands. Examples of such cues include: Shiny objects go into that box. Pull this way with respect to the object. Use this much force. This part needs to touch that part. When done, this part should not wiggle. Keep this upright. OLIVER will build upon modern machine learning techniques to exploit such cues for efficient, open-ended learning.
Innovation: Today's robots are either programmed for specific tasks, or can be trained to perform specific tasks. OLIVER seeks to bust these constraints by allowing robots to be taught diverse tasks, taking inspiration and methods from human teaching and learning. Beyond households, robots that can be taught in this way can be productively deployed in many different applications. OLIVER specifically addresses clean-up and recycling. This domain is of particular importance, as tidiness and sustainable use of resources immediately impact our quality of life.
Duration: 36 Months
“MICROVALU - Evaluating microbiodiversity in alpine pastures”
- Project coordinator: Eurac Research, Julia Seeber
- Project partner: Fondazione Edmund Mach, Heidi C. Hauffe
- Project partner: Paul Illmer, University of Innsbruck
When we think about habitats that are important for our wellbeing, we do not often think about soil. And yet, the soil allows us to cultivate crops, feed our livestock, grow trees for fuel and construction material and is the basis for a beautiful landscape and a valuable recreational area for locals and tourists. Even less known is that soil acts as a ‘carbon sink’ (buffering climate change), is responsible for detoxifying pollution, and gives us a clean water supply. Soil also provides a home for thousands of microbes and tiny animals, which keep the soil fertile and provide plants with nutrients. Despite soil being crucial to sustainable agriculture, very little is known about how soil, plant roots and soil organisms depend on one another to provide us with these services. Even less is known about the soil of grazing pastures, one of the most characteristic, and economically and culturally important Alpine habitats.
The project MICROVALU will bring together experts in soil ecology, microbiology, botany and zoology from across the Tyrol-South Tyrol-Trentino EUREGIO to try to understand how the microorganisms in all the components of the soil interact in Alpine pastures. These components are:
i) microorganisms in the soil that are responsible for decomposing organic matter and releasing nutrients;
ii) microorganisms in roots that release nutrients to plants;
iii) microorganisms in the guts of animals living in the soil who need microbes to help digest food;
iv) microorganisms in the faeces of animals living above the ground like beetles, rodents and livestock (cows or sheep) that also have a gut flora for digestion (the bacteria from their gut flora returns to the soil through defecation).
We will also test if these microorganisms change according to soil characteristics like acidity and humidity, to the diversity of animal communities themselves, or to altitude. The effect of altitude is particularly relevant because, by understanding how altitude changes the interactions in the soil, we can understand how climate might change the fertility of alpine pastures in the future. To complete this innovative project, we will take advantage of a long-term field site in the Province of Bolzano. In order to characterize the microbial communities of each soil component, we will use the laboratory technique ‘metataxonomics’. This new sequencing technique means all DNA can be extracted from each sample (soil, plant roots, worms, animal faeces, etc.) and for each sample, all of the microorganisms can be identified simultaneously. This work will be done in laboratories at the partner institutions where specialized equipment is already available. We will then combine all the data and use computer software to analyse the data to discover how the different soil components interact.
Duration: 36 Months
“Siderophore-iron uptake in E. amylovora and A. fumigatus”
- Project coordinator: Free University of Bozen-Bolzano, Stefano Benini
- Project partner: University of Trento, Sheref Mansy
- Project partner: Hubertus Haas, Medical University of Innsbruck
Towards new strategies in plant and human health. (SupErA) Plant and human pathogens utilize molecules called siderophores to obtain the iron that they need for survival from their host. We will study this mechanism and then use the information acquired to develop new therapies against plant and human pathogens, including one of the top ten bacterial plant pathogens. This bacterium, E. amyolovora, causes fire blight in Rosaceae (e.g., apple, pear). We will also develop therapies for A. fumigatus fungal infections, which causes a disease called aspergillosis in immune-compromised patients.
This proposal aims to decipher how E. amyolovora and A. fumigatus produce and absorb siderophores and to develop therapeutic molecules for plants and humans. This project employs an innovative interdisciplinary approach that exploits the expertise of Stefano Benini, structural biologist, Free University of Bozen-Bolzano, Sheref Mansy, biochemist, University of Trento and Hubertus Haas, molecular microbiologist, Medical University Innsbruck. First, the structure of the enzymes involved in the biosynthesis of siderophores will be determined at the molecular level. We will then reconstruct the pathway used by the investigated organisms to produce the siderophores, both in test tube and bacterial reactions. Finally, we will use the constructed pathways against the pathogens by rational modification to produce siderophore variants as therapeutic agents that inhibit fungal and bacterial growth. Further, new synthetic siderophores for medical imaging will be produced.
The project SupErA will lead to:
- A better comprehension of the biosynthesis and uptake of siderophores by bacteria and fungi.
- The development of biotechnologies for the production of molecules for the inhibition and/or imaging of bacterial and fungal pathogens in agriculture and medicine.
- The training of young researchers and divulgation of the results.
The creation of a network of research involving Tyrol, South Tyrol and Trentino This proposal brings together the three Euregio regions Tyrol, South Tyrol and Trentino in a joined effort that will strengthen the links between the institutions and increase their scientific visibility.
Duration: 36 Months
“Atmospheric boundary‐layer modeling over complex terrain”
- Project coordinator: Manuela Lehner, University of Innsbruck
- Project partner: Free University of Bozen-Bolzano, Massimo Tagliavini
- Project partner: University of Trento, Lorenzo Giovannini
The quality of weather forecasts strongly depends on the performance of numerical weather prediction models and areas of complex and mountainous terrain prove to be particularly challenging for numerical models. One of the reasons is that large spatial variations can occur within the atmospheric boundary layer, that is, the lowest layer of the atmosphere. The large heterogeneity of the terrain and differences in vegetation and land use at varying altitudes affect the partitioning of the solar radiation received at the surface and, subsequently, the transport of heat and moisture between the ground and the atmosphere. The transport of heat from the ground is, however, a major driving force for the temperature field within the boundary layer and spatial variations can lead to local wind circulations within mountainous terrain. A correct representation of the exchange of heat, moisture, and also momentum between the land surface and the atmosphere and within the boundary layer is thus crucial for successful model simulations. In numerical models this exchange between the land and the atmosphere and within the atmosphere is handled by the land surface model and the turbulence parameterization, respectively. Potential issues with respect to the performance of these parameterizations in mountainous terrain have been identified. For example, the parameterizations were generally developed based on observations from flat terrain and may thus not be entirely adequate for complex, mountainous terrain. Studies evaluating the performance of the parameterizations in mountainous terrain, however, are scarce.
The project “Atmospheric boundary-layer modeling over complex terrain” is a collaboration between the University of Innsbruck, the University of Trento, and the Free University of Bolzano with the objective to evaluate a model’s turbulence and land surface parameterizations in mountainous terrain and to identify potential issues with respect to these parameterizations that have a large impact on the model results. Improvements of these specific issues can thus be expected to have a comparatively large positive impact on numerical model simulations and consequently weather forecasts. To evaluate the model performance, model simulations are performed of individual case studies for two Alpine regions, the Austrian Inn Valley near Innsbruck, the Italian Adige Valley, and the mountain slopes near Bolzano. For both regions, several years of high-quality observational data are available for comparison with the model. The model’s sensitivity to changes or potential errors in the turbulence and land surface parameterizations is evaluated by means of idealized simulations, that is, certain parameters are modified systematically to quantify their impact on the model results.
Duration: 36 Months
“Comprehensive dissection of the superficial scald in apple”
- Project coordinator: Fondazione Edmund Mach, Fabrizio Costa
- Project partner: Laimburg Research Center, Angelo Zanella
- Project partner: Christian Huck, University of Innsbruck
The project entitled “Comprehensive dissection of the superficial scald disorder in apple through a multidisciplinary integration of OMICS approach” has been proposed to step forward into the physiological regulation of the superficial scald, a severe disorder occurring during the postharvest cold storage of apple. This physiological dysfunction generates a general discoloration of the fruit skin with the development of necrotic patches strongly impacting the final marketability. To protect the fruit from this phenomenon, effective strategies are nowadays available, such as the exogenous application of 1-methylcyclopropene (1-MCP), a molecule normally applied to delay the fruit ripening during storage. However, the application of this chemical compound is far from the required sustainability, especially in organic horticulture. As alternative, other strategies can be to date used, such as the storage at controlled atmosphere with low concentration of oxygen, although the physiological mode of action is still not fully understood. With this project we aim to unravel the underlying processes of the fruit protection mechanism. These results will improve our understanding about the effect of the storage technologies in preventing the appearance of the superficial scald and to ensure fruit safety and high quality standards through the development of more efficient and sustainable storage strategies. In this project, superficial scald will be also genetically investigated, developing, in the end, valuable molecular markers useful to support the selection of novel varieties characterized by higher postharvest performances. The methods employed will be therefore exploited on a full-sib population, made by crossing ‘Granny Smith’ and ‘Pinova’, showing either high susceptibility or resistance to superficial scald, respectively. The analysis of the variation of the metabolites across the individuals of the population, together with the detection of the DNA variants, will target the genomic regions most likely involved in the control of the scald protection. The most significant markers included in these intervals will represent novel tools to assist breeders in the selection phase. Despite the importance of the postharvest phase in apple, at present all the molecular markers developed and released to the scientific community are related primarily to aesthetic aspects and disease resistance. In the modern horticulture, however, the postharvest performance is a key factor contributing to the economic success of a cultivar. A superior storability would allow a more sustainable postharvest management, with the reduction of the chemical and energetic inputs currently applied. In the framework of this project, the several entities usually exploited singularly (such as physical, metabolic and transcriptomic assessment) will be integrated simultaneously into a unique effort. Furthermore, the analysis of expression QTLs (first case for apple) will lead towards a complete deciphering of the regulatory process leading to the scald protection in apple.
Duration: 36 Months
“Antiparasitic and anthelminitic mountain plants - HERBAL”
- Project coordinator: Fondazione Edmund Mach, Stefan Martens
- Project partner: Free University of Bozen-Bolzano, Matthias Gauly
- Project partner: Hermann Stuppner, University of Innsbruck
Parasitic infections of livestock animals, like cattle, sheep, or chicken are not only a burden for animals, but also for us consumers. One worm, for example, produces in the gut of an infected animal 70 000 eggs per day. Hence, worm infections are inevitable as soon as the animal is allowed to graze freely, which is standard on organic or small-scale farms. Commonly, infected animals are treated with a synthetic compound. The major concern is that the drug overuse create resistance, which occur in the process of evolution as Darwin it taught. Especially in worms that happens frequently, because the synthetic de-wormers have only one attack strategy. Realistically speaking since infections cannot be avoided especially in the free ranging chicken where the animals have a direct contact with their faeces the infection will occur again, in the worst case with a resistant worm. Furthermore, drug residues that entered the food chains can lead to cross-resistances in human worms. In sum, an urgent need for a new, long-lasting de-worming strategy exists.
As a solution the project HERBAL was invented. HERBAL aims to establish an organic de-worming strategy based on herbs which grow in the alpine region. This innovative project proposed by three leading researchers working in the Edmund Mach Foundation (Trentino, FEM, Biotechnology of Natural Product group – Stefan Martens; coordinator), the Free University of Bolzano (South Tyrol, Animal Science – Matthias Gauly), and the University of Innsbruck (Tyrol, Pharmacognosy – Hermann Stuppner), respectively. Together they plan to study ten herbs, which were selected based on the cultural knowledge of regional farmers. The researchers believe that these herbs contain many compounds which kill worms, thus a herbal extract will effectively attack the worm making the deveoloment of resistance difficult. In the beginning, to show that herb extract work, they will test them against worm-eggs and select the three best extracts. Later, they will feed infected chicken to treat them against the worms. The effectiveness of the extract will be monitored by counting the number of worm eggs in the faeces. Furthermore, the researchers assume, that production traits – for example the number of eggs laid by the hens – are not affected or positively influenced, when is supplemented in the diet of the chickens with the herbal extract. Additionally, the researcher aim to discover the molecular principals behind the activity. Based on this data, and with modern computational techniques they will analyze the activity in depth, as they belief that even more active natural compounds exists.
Duration: 36 Months
“Roles and targeting of RIOK1 in resistant prostate cancer”
- Project coordinator: Zoran Culig, Medical University of Innsbruck
- Project partner: University of Trento, Peter De Wulf
While the kinase RioK1 is essential for the viability and survival of any organism, its biological activity is not well understood. RioK1 is produced in abnormally high amounts in many cancers, strongly suggesting that elevated levels of the kinase contribute to cancer formation, growth and metastasis. Indeed, recent reports confirmed this hypothesis. This validation now adds major importance onto (1) determining the biology of RioK1, (2) evaluating RioK1 as an anti-cancer drug target, and (3) promoting the kinase as a candidate cancer biomarker.
In this research project, the lab of Peter De Wulf (Trento University) will focus on studying RioK1 biology and defining at the molecular level its roles inside a cell. Specifically, RioK1 will be eliminated genetically or inhibited (using the drug toyocamycin) to map its activity at the protein, RNA and DNA levels allowing us to computationally combine these data into a functional network. We will experimentally corroborate it and measure how RioK1 overexpression influences the network possibly resulting in cancerous transformation. We will further probe whether inhibiting its pathological activity with toyocamycin can eradicate cancer cells. If so, this drug can then be further evaluated (outside the scope of this grant) in future clinical trials.
Our project also has a translational dimension to it (performed by Zoran Culig’s lab, Innsbruck University), which is focused on prostate cancer. RIOK1, overexpressed in >5% of all prostate cancers, has been linked to the morphological changes required for tumor cancer (the “epithelial-to-mesenchymal transition”) and metastasis of prostate cancer, and other malignancies. Prostate cancer is one of the most commonly detected tumors and can be cured if identified in its earliest stages. Disease-advanced patients receive endocrine- or chemotherapy to which resistance inevitably develops, resulting in death. Prostate cancer models of chemotherapy resistance are available in the Culig lab and will be used intensively for the project. We will also probe whether RioK1 activity contributes to the development of resistance and address if toyocamycin can act as an anti-tumorigenic drug in prostate cancer cells that have developed resistance to current therapies including to the anti-androgen medicine enzalutamide. Depending on the results of these experiments, the lab will advance these studies in mice, using xenograft models.
In summary, combining basic and translational studies of RioK1 activity will reveal for the first time ever its contribution to cancer biology and the development of chemotherapy resistance. Examining the anticancer effects of inhibiting the kinase may provide a new, potent tool to eradicate prostate (and other) cancers that are untreatable today.
Duration: 36 Months
Project rated B (C1) “very good” funded:
“Social Impact of Collective Ownership in the Alpine Areas”
Resubmission of IPN 41
- Project coordinator: University of Trento, Ericka Costa
- Project partner: Free University of Bozen-Bolzano, Massimiliano Bonacchi
- Project partner: Oliver Bender, Austrian Academy of Sciences
This project aims to investigate the social and environmental impact of collective ownerships in the Alpine areas of the EGTC (European Grouping of Territorial Cooperation) of Trentino, South Tyrol and Tyrol. These collective ownerships manage the environment in different Alpine areas and countries both for pastured and forested areas. They provide services to the owners and communities while preserving the environment in which they are located.
This type of collective ownership manages and contributes to generating a multifaceted type of capital, simultaneously natural, cultural and economic. Indeed, these collectives represent an asset to be transmitted within the family from parents to children, where conservative and productive use of assets both exist.
These organisations have always been not-for-profit and based on a sense of belonging and mutual aid. For people living in the Alps, these forms of collective ownership represent another way of partially owning the territory in which they live because they helped (and still help) people living in mountain areas to manage their territory and other assets accumulated during their centenary existence. In recent years, however, some problems have emerged: the sense of belonging characterising Alpine collective ownership seems to be gradually waning, causing problems in the effectiveness of their environmental actions.
In light of the aforementioned problems, this project aims to measure the social and environmental impact of such organisations in order to deepen their current influence on local communities. The idea is that by enabling these institutions to overcome possible problems connected to the consensus regarding their activities, they can still provide a valuable contribution to sustainably managing natural resources and to aggregate individuals on Alpine lands.
Ultimately, the project proposes instruments for measuring collective ownerships’ social and environmental impacts while evaluating how possible problems related to their identity can influence their actions regarding the environment they must preserve and develop. The necessity to measure social impact arises from the peculiarities of collective ownerships, organisations for which traditional financial performance analyses provide scant informative value because they are not driven by the profit maximisation paradigm. The importance and strong interest in measuring these impacts are demonstrated by a recent policy brief by the Organisation for Economic Co-operation and Development of the European Commission.
The researchers will make use of adequate communication channels and take into account the cultural, linguistic and historical features of the territory.
Duration: 24 Months